@article {IOPORT.05308324,
    author       = {Sun, Mingxuan},
    title        = {Iterative learning control with initial state learning.},
    year         = {2007},
    journal      = {Control and Decision},
    volume       = {22},
    number       = {8},
    issn         = {1001-0920},
    pages        = {848-852},
    publisher    = {Editorial Office of Control and Decision, Northeastern University, Shenyang},
    abstract     = {Summary: By novel initial state learning, the assumption on initial repositioning is relaxed for the conventional iterative learning control. It is usually assumed that at the beginning of each trial, the initial state is reset to the desired one without repositioning errors. The learning schemes under consideration are of robust convergence, which allow initial repositioning errors and initial states not to be specified positions. Sufficient conditions for the convergence of second-order LTI system, by which learning gains can be chosen, are given. The learning schemes can overcome imperfect knowledge about system dynamics to achieve complete tracking, though the initial state learning laws are independent of the input matrix.},
    identifier   = {05308324},
}